On 09/12/04 23:42 +0100, Frank Ellermann wrote:
Devdas Bhagat wrote:
What about using multiple .forwards or equivalent?
Forwarding is something the receiver does. Sender policies are
something the senders publish. Senders don't know and don't
care what a receiver does, it's his problem. Senders only say:
Right. But .forward is still used and SPF breaks that. SRS is a hack to
try and prevent it breaking.
"These are the IPs you as 3rd party see in MAIL FROM me to you
at your MX, any other mail claiming to be MAIL FROM me at your
MX is forged, don't bother to accept it, please reject it."
Why's that so difficult to understand ? Senders have no idea
how a receiver forwards mail behind his MX, it's none of their
business. The receiver is free to do whatever he likes. And
if his mail setup is broken he owns the pieces.
Breaking a known working feature is not a good way to keep those
power users happy.
Or people using the bounce feature in mutt?
Is that the stuff where I get mail claiming to be MAIL FROM me,
but in fact it's from somebody else ? I'd report it to abuse@
for the corresponding IP, it's forged. But actually it should
never arrive if publishers of -all policies also check SPF on
their own MX.
No. Mutt adds a Resent-from header and keeps the original mail
information. This needs deliberate action by the sender, and is the
equivalent of a .forward (slightly loosely speaking) for that mail.
Devdas Bhagat
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg